Unassigned Devices - Managing Disk Drives and Remote Shares Outside of The Unraid Array


Recommended Posts

I'm having an issue where my drive keep unmounting. I attached a drive so that I can copy data from the server but when I check on it it cannot write to the device and it had been unmounted. I don't see anything in the logs indicating it was unmounted.

 

/dev/sdc:
re-reading partition table
May 09 12:02:49 Creating a 'gpt' partition table on disk '/dev/sdc'.
May 09 12:02:49 Creating a primary partition on disk '/dev/sdc'.
May 09 12:02:49 Formatting disk '/dev/sdc' with 'ext4' filesystem.
May 09 12:02:50 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 12:25:38 Format disk '/dev/sdc' with 'ext4' filesystem result:
Creating filesystem with 1220942336 4k blocks and 152621056 inodes
Filesystem UUID: 87a2afe6-4be3-4da1-bcbf-5fbd006a09d3
Superblock backups stored on blocks: 
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 
4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968, 
102400000, 214990848, 512000000, 550731776, 644972544

Allocating group tables: 0/3done 
Writing inode tables: 0/3done 
Creating journal (262144 blocks): done
Writing superblocks and filesystem accounting information: 0/3done
May 09 12:25:41 Reloading disk '/dev/sdc' partition table.
May 09 12:25:42 Reload partition table result:
/dev/sdc:
re-reading partition table
May 09 12:25:42 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 12:35:19 Adding disk '/dev/sdc1'...
May 09 12:35:19 Mount drive command: /sbin/mount -t ext4 -o auto,noatime,nodiratime,async,nodev,nosuid '/dev/sdc1' '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'
May 09 12:35:20 Successfully mounted '/dev/sdc1' on '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'.
May 09 12:35:20 Partition 'TOSHIBA HDWE150 272BK5OOF57D' is not set as sharable and will not be shared...
May 09 12:35:20 Device '/dev/sdc1' script file not found. 'ADD' script not executed.
May 09 15:27:35 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 16:29:47 Adding disk '/dev/sdp1'...
May 09 16:29:47 Mount drive command: /sbin/mount -t ext4 -o auto,noatime,nodiratime,async,nodev,nosuid '/dev/sdp1' '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'
May 09 16:29:57 Successfully mounted '/dev/sdp1' on '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'.
May 09 16:29:57 Partition 'TOSHIBA HDWE150 272BK5OOF57D' is not set as sharable and will not be shared...
May 09 16:29:57 Device '/dev/sdp1' script file not found. 'ADD' script not executed.
May 09 17:58:13 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 19:44:22 Adding disk '/dev/sdq1'...
May 09 19:44:22 Mount drive command: /sbin/mount -t ext4 -o auto,noatime,nodiratime,async,nodev,nosuid '/dev/sdq1' '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'
May 09 19:44:34 Successfully mounted '/dev/sdq1' on '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'.
May 09 19:44:34 Partition 'TOSHIBA HDWE150 272BK5OOF57D' is not set as sharable and will not be shared...
May 09 19:44:34 Device '/dev/sdq1' script file not found. 'ADD' script not executed.

 

Link to comment
4 minutes ago, bobbintb said:

I'm having an issue where my drive keep unmounting. I attached a drive so that I can copy data from the server but when I check on it it cannot write to the device and it had been unmounted. I don't see anything in the logs indicating it was unmounted.

 


/dev/sdc:
re-reading partition table
May 09 12:02:49 Creating a 'gpt' partition table on disk '/dev/sdc'.
May 09 12:02:49 Creating a primary partition on disk '/dev/sdc'.
May 09 12:02:49 Formatting disk '/dev/sdc' with 'ext4' filesystem.
May 09 12:02:50 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 12:25:38 Format disk '/dev/sdc' with 'ext4' filesystem result:
Creating filesystem with 1220942336 4k blocks and 152621056 inodes
Filesystem UUID: 87a2afe6-4be3-4da1-bcbf-5fbd006a09d3
Superblock backups stored on blocks: 
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 
4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968, 
102400000, 214990848, 512000000, 550731776, 644972544

Allocating group tables: 0/3done 
Writing inode tables: 0/3done 
Creating journal (262144 blocks): done
Writing superblocks and filesystem accounting information: 0/3done
May 09 12:25:41 Reloading disk '/dev/sdc' partition table.
May 09 12:25:42 Reload partition table result:
/dev/sdc:
re-reading partition table
May 09 12:25:42 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 12:35:19 Adding disk '/dev/sdc1'...
May 09 12:35:19 Mount drive command: /sbin/mount -t ext4 -o auto,noatime,nodiratime,async,nodev,nosuid '/dev/sdc1' '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'
May 09 12:35:20 Successfully mounted '/dev/sdc1' on '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'.
May 09 12:35:20 Partition 'TOSHIBA HDWE150 272BK5OOF57D' is not set as sharable and will not be shared...
May 09 12:35:20 Device '/dev/sdc1' script file not found. 'ADD' script not executed.
May 09 15:27:35 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 16:29:47 Adding disk '/dev/sdp1'...
May 09 16:29:47 Mount drive command: /sbin/mount -t ext4 -o auto,noatime,nodiratime,async,nodev,nosuid '/dev/sdp1' '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'
May 09 16:29:57 Successfully mounted '/dev/sdp1' on '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'.
May 09 16:29:57 Partition 'TOSHIBA HDWE150 272BK5OOF57D' is not set as sharable and will not be shared...
May 09 16:29:57 Device '/dev/sdp1' script file not found. 'ADD' script not executed.
May 09 17:58:13 Disk with serial 'TOSHIBA_HDWE150_272BK5OOF57D' is not set to auto mount and will not be mounted...
May 09 19:44:22 Adding disk '/dev/sdq1'...
May 09 19:44:22 Mount drive command: /sbin/mount -t ext4 -o auto,noatime,nodiratime,async,nodev,nosuid '/dev/sdq1' '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'
May 09 19:44:34 Successfully mounted '/dev/sdq1' on '/mnt/disks/TOSHIBA_HDWE150_272BK5OOF57D'.
May 09 19:44:34 Partition 'TOSHIBA HDWE150 272BK5OOF57D' is not set as sharable and will not be shared...
May 09 19:44:34 Device '/dev/sdq1' script file not found. 'ADD' script not executed.

 

See if the syslog has any entries that would indicate an unmount.

Link to comment
10 hours ago, bobbintb said:

Thanks, I thought I checked before but I guess I didn't look back far enough. I'm getting a bunch of write errors. I guess I should run pre-clear a few more times. Might have a DOA.

Write errors would be a problem.  Once UD has mounted a drive, any issues with reading, writing, and un-mounting would be logged in the syslog and not in the UD log.

  • Upvote 1
Link to comment

Having issue when mounting more than one disk. The 2nd disk takes over.

 

 

unRAID server Pro, version 6.3.3

Unassigned.devices 2017.05.04a

 

Disk2: 9650SE-12M_DISK_9VK0FZTY000000000000

Disk1: 9650SE-12M_DISK_A8346038000000000000

 

If one or both are mounted, they show up as below, I think this might be the root of my problem. Could it be the 3ware controller I'm using in Jbod?

/mnt/disks/AMCC_9650SE_12M_DISK

This is even if i use the + and change the name, wi will always default back, and when I go to rename again, the name I changed it to is still there.

 

	9650SE-12M_DISK_A8346038000000000000*	xfs	2 TB	0 	32.2 GB	1.97 TB
		`->1/mnt/disks/AMCC_9650SE_12M_DISK

I have a feeling I'm missing something stupid here, but any input is helpful.

 

 

 

 

 

Link to comment
49 minutes ago, The_Doctor said:

Having issue when mounting more than one disk. The 2nd disk takes over.

 

 

unRAID server Pro, version 6.3.3

Unassigned.devices 2017.05.04a

 

Disk2: 9650SE-12M_DISK_9VK0FZTY000000000000

Disk1: 9650SE-12M_DISK_A8346038000000000000

 

If one or both are mounted, they show up as below, I think this might be the root of my problem. Could it be the 3ware controller I'm using in Jbod?


/mnt/disks/AMCC_9650SE_12M_DISK

This is even if i use the + and change the name, wi will always default back, and when I go to rename again, the name I changed it to is still there.

 


	9650SE-12M_DISK_A8346038000000000000*	xfs	2 TB	0 	32.2 GB	1.97 TB
		`->1/mnt/disks/AMCC_9650SE_12M_DISK

I have a feeling I'm missing something stupid here, but any input is helpful.

 

 

 

 

 

- Remove both disks and then delete them from the historical devices.

- Verify the /mnt/disks/AMCC_9650SE_12M_DISK mount point is not there.  If it is unmount it and then rm -rf /mnt/disks/AMCC_9650SE_12M_DISK.  If you can't clear it up, reboot the server.

- Install one disk and change the mount point.  Remember to hit <Return> to save the new mount point name.

- Mount this disk and be sure the mount point is what you set.  If not the config file is probably corrupted.

- If this works add the second disk and change the mount point.

 

If this doesn't work, post your /flash/connfig/plugins/unassigned.devices/unassigned.devices.cfg file and let me see what it contains.

 

  • Upvote 1
Link to comment
5 minutes ago, dlandon said:

- Remove both disks and then delete them from the historical devices.

- Verify the /mnt/disks/AMCC_9650SE_12M_DISK mount point is not there.  If it is unmount it and then rm -rf /mnt/disks/AMCC_9650SE_12M_DISK.  If you can't clear it up, reboot the server.

- Install one disk and change the mount point.  Remember to hit <Return> to save the new mount point name.

- Mount this disk and be sure the mount point is what you set.  If not the config file is probably corrupted.

- If this works add the second disk and change the mount point.

 

If this doesn't work, post your /flash/connfig/plugins/unassigned.devices/unassigned.devices.cfg file and let me see what it contains.

 

 

I must have missed this part when reading over the guide. I never hit <return> When changing the mount point name. 

 

Thank you for a quick reply, and fix!

Link to comment
9 minutes ago, dlandon said:

The 'Help' button is your friend.

Just an FYI, the having to hit <return> is not listed under help info on the GUI. It is however, on your first post

 

Help Info from GUI:

"Additional options are available when you click on the device identification. When the disk is unmounted, the mount name of the device can be changed. A disk can be formatted and an fsck run to check the disk."

 

From your first post:

"There are different operations in various active areas (especially the disk serial number) based on whether or not the disk drive is mounted.  If the disk is not mounted, click on the '+ icon by the serial number, click on the partition name, then you can type a new mount point name.  Press Enter for the name change to be applied.  This will now become the mount point and the share name when the disk is shared."

 

This is still on me, for missing something in your guide, but thought I would share in case this can be updated moving forward.

 

Link to comment
31 minutes ago, The_Doctor said:

Just an FYI, the having to hit <return> is not listed under help info on the GUI. It is however, on your first post

 

Help Info from GUI:

"Additional options are available when you click on the device identification. When the disk is unmounted, the mount name of the device can be changed. A disk can be formatted and an fsck run to check the disk."

 

From your first post:

"There are different operations in various active areas (especially the disk serial number) based on whether or not the disk drive is mounted.  If the disk is not mounted, click on the '+ icon by the serial number, click on the partition name, then you can type a new mount point name.  Press Enter for the name change to be applied.  This will now become the mount point and the share name when the disk is shared."

 

This is still on me, for missing something in your guide, but thought I would share in case this can be updated moving forward.

 

That can be fixed.

  • Upvote 1
Link to comment
1 hour ago, The_Doctor said:

Just an FYI, the having to hit <return> is not listed under help info on the GUI. It is however, on your first post

 

Help Info from GUI:

"Additional options are available when you click on the device identification. When the disk is unmounted, the mount name of the device can be changed. A disk can be formatted and an fsck run to check the disk."

 

From your first post:

"There are different operations in various active areas (especially the disk serial number) based on whether or not the disk drive is mounted.  If the disk is not mounted, click on the '+ icon by the serial number, click on the partition name, then you can type a new mount point name.  Press Enter for the name change to be applied.  This will now become the mount point and the share name when the disk is shared."

 

This is still on me, for missing something in your guide, but thought I would share in case this can be updated moving forward.

 

Fixed.  I added additional text to the help and tool tip.  Help is your friend again!

Edited by dlandon
  • Upvote 1
Link to comment
On 9.5.2017 at 0:26 PM, dlandon said:

I cannot reproduce the issue you are seeing with UD disk spin ups.  Post your diagnostics for me to look at.

I can confirm the behaviour amigenius is observing. If notifications are enabled the UD disks do no longer spin-down. Most likely in order to generate notifications unraid polls all drives (e.g. for temperature) in certain intervals. While unraid correctly handles the array drives (maybe only polling while spinning) and other drives if UD is not installed, it seems that with UD installed and UD shareing  of disks the polling logic does no longer work when notifications are enabled.

Edited by Videodr0me
Link to comment
3 hours ago, Videodr0me said:

I can confirm the behaviour amigenius is observing. If notifications are enabled the UD disks do no longer spin-down. Most likely in order to generate notifications unraid polls all drives (e.g. for temperature) in certain intervals. While unraid correctly handles the array drives (maybe only polling while spinning) and other drives if UD is not installed, it seems that with UD installed and UD shareing  of disks the polling logic does no longer work when notifications are enabled.

What kind of drives?

Link to comment

You have Seagate drives.  It seems to be related to Seagate drives.  The only difference in UD being installed or not installed is that with UD they can be mounted and the temperature is checked when the UD webui is active.  Checking the temperature on some drives causes them to spin up.  UD does nothing to the drives that is related to SMART except to set a sleep timer on the drive.

 

What happens when UD is installed and the disks are not mounted?  What happens when UD is installed and the disks are mounted?

Link to comment
43 minutes ago, zin105 said:

Is it possible to run a daily script that mounts a network share, runs a rsync backup, and unmounts as soon as it's done? Want to minimze the time the backup destination is mounted in case of ransomware.

Yes.  Use the User Scripts plugin and set up a script on a schedule to mount the device, run your script, and unmount the device.  The downside to this is that you need to mount/unmount using the /dev/sdX designation and it can change on each boot.  You could add the logic to look up the /dev/sdX for the device and then use that to mount and unmount it.

 

The UD script to perform mounting and unmounting is documented in the UD plugin help.

 

EDIT: A better way is to leave it mounted and limit access or set limited permissions with SMB.  That way it can't be written to.

Edited by dlandon
Link to comment
46 minutes ago, zin105 said:

 

Could you elaborate on this part? I don't understand how I would give permission enough to write the backup, but not enough that a ransomware can.

Using rsync does not require SMB write access.  Ransomware cannot access the drive except through SMB.

Edited by dlandon
Link to comment
On 15.5.2017 at 1:22 PM, dlandon said:

You have Seagate drives.  It seems to be related to Seagate drives.  The only difference in UD being installed or not installed is that with UD they can be mounted and the temperature is checked when the UD webui is active.  Checking the temperature on some drives causes them to spin up.  UD does nothing to the drives that is related to SMART except to set a sleep timer on the drive.

 

It is highly unlikely that it has anything to do with the seagate drives:

1) All drives are kept spinning not only the seagate drives

2) Without UD all drives do spin down, even when looking at the dashboard in unraid webui - and even with notifications enabled

3) With UD installed and notifications disabled, all drives spin down properly EVEN when having the dashboard in the unraid web ui open AND EVEN when having the UD devices TAB open

 

 

But do not worry to much, i am not using notifications - I just tried to confirm amigenius findings, and surprisingly my system behaves the same as his. This behavior seems to have been introduced after 6.1.9, because with 6.1.9 it worked (as far as i remember, as i used notifications back then). Its no biggie though as it is not really affecting UD operation.

Link to comment

I cannot reproduce the problem and can therefore not make any kind of fix.  I have all WD drives and do not have this problem.  No one is giving me anything definitive that I can use to try to find the problem.  Just saying 'It's broken' gives me nothing to work on to find the problem.  Are the UD drives mounted or unmounted?  Is there a UD script on the drive that is spinning up?  What is the controller the drive is connected to?

Link to comment

I'm not sure how any of you with the disk spin up issue are creating this situation with UD installed and uninstalled and determining that it is a UD problem.  I have a drive mounted in a docking station in my test system and I was able to make this happen.  It happens with UD uninstalled.  The problem appears to be in the background monitor script built into unRAID.  This should be posted as a defect in unRAID as I can make it happen without UD being installed and monitoring enabled.

  • Upvote 1
Link to comment

I upgraded to 6.3.4; UD is upgraded to 2017.5.21;  I put in a new WD red 4tb disk. it shows up in UD, I can format it and mount it, but I am unable to use preclear. When i click on preclear nothing happens. I have used UD in the past to preclear disks. 

I uninstalled and reinstalled to make sure that there wasn't a corrupt file from upgrading. 

Just checking to see if there is something wrong on my end or if the upgrade to 6.3.4 broke something.

 

Link to comment
  • trurl pinned this topic

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.